Systematic screening of transition metal dual-atom-doped phthalocyanine electrocatalysts for the oxygen reduction reaction

Abstract

Electrocatalysis is emerging as a promising technology for dealing with the energy crisis and environmental challenges. Phthalocyanines doped with two transition metal atoms (Tm₂-Pc) show outstanding ORR catalytic activity and offer several catalytic sites. However, a systematic investigation of the ORR catalytic efficacy of Tm₂-Pc monolayers with varying central metal atoms is lacking, hindering the establishment of design principles for this family of materials. Our research findings demonstrate that the d-band center, HOMO–LUMO, and adsorption energy are significant factors in the evaluation of catalysts. The ORR efficacy of Tm₂-Pc is dependent on the interaction strength between the adsorbate and Tm₂-Pc. With overpotentials of just 0.28 eV and 0.27 eV, respectively, Mn₂-Pc and Ru₂-Pc show good stability and outstanding catalytic activity. This work provides a theoretical foundation for the creation of high-performance catalysts by revealing the ORR catalytic mechanism of Tm₂-Pc.

---

1. Introduction

The persistent increase in global energy demand, coupled with environmental damage caused by fossil fuels, has made the search for sustainable and clean energy alternatives a central focus of contemporary scientific research.^1–3 Clean batteries are a significant method for mitigating environmental pollution.^4,5 Nonetheless, identifying economical and consistently high-performing ORR catalytic materials for batteries continues to pose a considerable difficulty. While phthalocyanines doped with single transition metal atoms demonstrate commendable catalytic capability, their catalytic efficiency requires further examination.^6,7 In contrast to single transition atom-doped Pc, Tm₂-Pc offers a greater number of catalytic sites, thereby enhancing the ORR catalytic performance.^8,9 The interactions and catalytic processes of Tm₂-Pc in the ORR process are inadequately comprehended. Consequently, a methodical examination of the ORR catalysis of Tm₂-Pc is crucial.

To systematically elucidate the interactions and catalytic processes of Tm₂-Pc in the ORR, one strategy is traditional experimental techniques.^10–12 This approach initially synthesizes the Tm₂-Pc materials and thereafter evaluates their ORR catalytic performance. Thoroughly evaluating the catalytic efficacy of many materials individually is time-consuming and yields data that are affected by experimental conditions, complicating the elucidation of the catalytic mechanism. An additional method is systematic screening that is performed using first principles.^13–15 First-principles approaches provide microscopic insight into the interactions and catalytic processes in the oxygen reduction reaction. Simultaneously, systematic screening identifies the Tm₂-Pc materials as exhibiting optimal ORR catalytic performance among a multitude of candidate materials. This approach offers a theoretical foundation for traditional tests while markedly reducing both time and cost.

Several pertinent studies have emerged about the catalytic efficacy of phthalocyanines doped with double transition metal center atoms.^16–19 Wang et al.²⁰ designed 18 varieties of heteronuclear diatomic Pc catalysts, of which FeTi@Pc has a limiting potential of −0.18 eV, enabling the reaction to proceed spontaneously. Mirshokraee et al.²¹ synthesized an iron-based Pc catalyst, revealing that the overpotential for the ORR is 0.93 eV. The findings by Ma et al.⁹ demonstrated that the Fe@Pc/Co@Pc substrates display superior catalytic performance for the ORR. Liao et al.⁸ synthesized Fe@Pc, Ni@Pc, and FeNi@Pc substrates, with the overpotential for the ORR reaction of FeNi@Pc measured at 0.85 eV. The aforementioned research results demonstrate that phthalocyanine doped with dual transition metal atoms displays superior electrocatalytic efficacy for the ORR. The transition state atoms have only been partially studied in the current studies, and it is as yet unknown how Tm₂Pc interacts and catalyzes reactions.

This work utilizes systematic screening grounded in first principles to comprehensively evaluate prospective Tm₂-Pcs, meticulously examining the interaction dynamics between Tm and Pc, and elucidating the catalytic processes of Tm₂-Pc. The initial screening of the catalytic substrates was performed by evaluating the formation energy, adsorption energy, d-band center, density of states, and the ab initio molecular dynamics (AIMD) simulations. Subsequently, the catalytic mechanism of Tm₂-Pc was examined through the examination of the reaction pathway and the volcano plot. Ultimately, the electronic interaction patterns of Tm₂-Pc were revealed through Crystal Occupation Hamiltonian Population (COHP), differential charge density, molecular orbitals, and Bader charge analysis.

2. Computational method

The Vienna Ab initio Simulation Package (VASP), renowned for its efficacy in modeling catalytic materials, was utilized to conduct Density Functional Theory (DFT) calculations.^22,23 The Perdew–Burke–Ernzerhof (PBE) variant of the generalized gradient approximation (GGA) was used to characterize the exchange–correlation function. Utilizing the projector-augmented wave method (PAW), the interaction between valence electrons and ions was described.^24–27 Spin polarization was considered, and the van der Waals (vdW) interaction was corrected.²⁸ The Brillouin region was sampled using a gamma-centered 3 × 3 × 1 k-point grid.²⁹ For geometry optimizations, the convergence standard of energy and force was set to 10⁻⁶ eV and 0.01 eV Å⁻¹ per atom, and a cutoff energy of 520 eV was applied for the plane-wave basis set.³⁰ The thermal stability of the catalyst candidates was assessed using AIMD simulations conducted at 300 K for 5 ps, with a time step of 1 fs and temperature regulation implemented via the Nosé–Hoover technique.^23,31 The COHP method, which is implemented by LOBSTER, was employed to analyze the electronic interaction mechanism.^32,33 A Bader charge analysis was employed to quantitatively characterize the transfer of charge between the substrate and the adsorbates.^34,35

In acidic solution, the ORR occurs via a four-electron transfer mechanism, as illustrated below:^35–37

* + O₂(g) + H⁺ + e⁻ → OOH* (1)

OOH* + H⁺ + e⁻ → O* + H₂O(l) (2)

O* + H⁺ + e⁻ → OH* (3)

OH* + H⁺ + e⁻ → H₂O(l) + * (4)

The asterisk (*) denotes an active site on the catalyst surface, whereas O*, OH*, and OOH* represent the adsorbed intermediates. (l) and (g) imply the liquid and gas phases, respectively. The variation in free energy (ΔG) at each stage is quantified by the subsequent equation:³⁸

ΔG = ΔE + ΔE_ZPE − TΔS (5)

where ΔE signifies the energy disparity between the reactant and product, while ΔE_ZPE and ΔS indicate the alterations in the zero-point energy difference and the entropy contribution of the geometry, respectively, derived from the computed vibrational frequencies and standard tables of reactants and products. The overpotential for the ORR (η_ORR) is then computed to assess its performance:³⁹where ΔG₁, ΔG₂, ΔG₃ and ΔG₄ are the free energy changes of the basic ORR step (1–4).

The adsorption energy (E_ads) of O₂ on the transition metal atoms of the catalyst is a measure for assessing ability to adsorb. The E_ads is represented by the subsequent formula:⁴⁰

E_ads = E_*O2 − E_O2 − E_Tm2-Pc (7)

where E_*O2, E_O2, and E_Tm2-Pc are the energies of the Tm₂-Pc with O₂, O₂ gas, and Tm₂-Pc substances, respectively.

The formation energy (E_form) of Tm-Pc is displayed by the subsequent formula:⁴¹

E_form = E_Tm2-Pc − n × E_Tm − E_Pc (8)

where E_Tm2-Pc, E_Tm, and E_Pc are the energies of Tm₂-Pc, the total energy of the metal atom in its most stable bulk structure,²⁰ and Pc, respectively. n is the number of Tm atoms.

The dissolution potential (U_diss) of Tm-Pc is defined as follows:^42,43

where U^o_diss and ne are the standard dissolution potential and the number of electrons of the Tm, respectively.

3. Results and discussion

3.1 High-throughput screening method

To identify high-performance and structurally stable catalysts, the catalytic effectiveness of 29 different Tm₂-Pc structures was systematically evaluated. At the outset, Tm₂-Pc compounds exhibiting superior ORR catalytic efficacy were identified, and their structural stability and O₂ chemisorption were examined. Secondly, the electronic property of Tm₂-Pc was assessed, followed by an examination of the catalytic performance, as illustrated in Fig. 1. Prior to screening Tm₂-Pc, carcinogenic and radioactive transition elements, including Hg, Tc, Cd, and others, were first rejected.^44,45 Furthermore, rare earth metals were omitted owing to the challenges associated with extraction and significant environmental contamination.⁴⁶ Consequently, 26 transition elements were predominantly chosen as the atom center (Fig. 1).

Fig. 1 Screening process and candidate TM materials: (a) scheme of the workflow for high-throughput, and (b) TM candidate elements of TM-Pc.

3.1.1 Screening method. This study at first integrated dual transition metal-doped phthalocyanines to create a Tm₂-Pc structure, as seen in Fig. S1. The formation energy serves as an indicator of the structural integrity of Tm₂-Pc. Table S1 exhibits that the formation energy of Tm₂-Pc is negative, signifying the stability of the Tm₂-Pc structure. Furthermore, the dissolution potential is positive (Table S2), and both the formation energy and dissolution potential indicate the stability of Tm₂-Pc in terms of thermal and electrochemical stability, respectively. The essential factor for the four-electron ORR is the adsorption of O₂ onto the substrate, which is necessary for the reaction to occur efficiently. The O₂ adsorption energy on Zn-Pc is greater than zero, whereas it is less than zero on the other Tm₂-Pc (Fig. 2(a)). O₂ cannot adsorb on the substrate if the adsorption energy is larger than zero, which rules out Zn₂-Pc. Simultaneously, an extremely elevated adsorption energy is detrimental to the generation of intermediates, including *OOH, *O, and *OH. The current study findings demonstrate that the adsorption energy of O₂ on the adsorbent is below 0.8 eV,^47–49 facilitating the smooth progression of the ORR; thus, Tm₂-Pc with elevated adsorption energy is dismissed. Simultaneously, the d-band center and the HOMO–LUMO band gap are employed to assess the interaction intensity and effectiveness for electron transitions throughout the catalytic process, respectively. The proximity of the d-band center to the Fermi level enhances O₂ adsorption and enables O–O bond breakage; conversely, a greater distance from the Fermi level diminishes adsorption and impedes the performance of the ORR.^50,51 Upon analyzing the adsorption energy and d-band center, Cu₂-Pc, Ag₂-Pc, and Au₂-Pc are omitted.

Fig. 2 Interaction rules of Tm₂-Pc for the ORR: (a) adsorption energy of the candidate Tm₂-Pc, and (b) d-band center and HOMO–LUMO of the candidate Tm₂-Pc.

A critical consideration is that when the d-band center exceeds zero, its effectiveness to donate electrons to the adsorbate diminishes, which is detrimental to the ORR catalytic process. Consequently, Tm₂-Pc (where Tm represents Sc, Ti, V, Y, Zr, Hf, W) is ignored (Fig. 2(b)). A lower HOMO–LUMO band gap facilitates electron transfer, whereas a larger band gap hinders the reaction,^52–54 thereby excluding Tm (Tm = Ni, Rh, Pd, Os, Ir, Pt)-Pc. The primary focus is on four candidate catalysts, namely Tm (Tm = Mn, Fe, Co, Ru)-Pc, which are excluded due to their unfavorable performance in promoting the ORR, as determined from the adsorption energy, d-band center, and HOMO–LUMO analysis of the Tm₂-Pc.

3.1.2 Screening the stability of the Tm₂-Pc. The stability of the candidate catalysts is a critical requirement for the ORR to proceed. Dynamic and binding stabilities are examined using AIMD and projected density of states (PDOS), respectively. An explanation of the dynamic stability of the structure is provided by AIMD at the atomic and electronic scales. Fig. 3 illustrates the AIMD of Tm (where Tm represents Mn, Fe, Co, or Ru) at 300 K. Tm-Pc exhibits no notable alterations in energy or structure. These structures are concurrently examined by AIMD at an elevated temperature of 1000 K (Fig. S2). Significant changes in energy and temperature have yet to be observed, and structural changes are negligible. AIMD investigations demonstrate that these catalysts preserve structural stability at both ambient and elevated temperatures.

Fig. 3 The stability evaluation of Tm₂-Pc: (a)–(d) PDOS and AIMD of Mn₂-Pc, Fe₂-Pc, Co₂-Pc and Ru₂-Pc.

Electron transfer and orbital hybridization are the primary mechanisms that control the binding stability between O₂ and Tm₂-Pc. The PDOS of the chosen Tm₂-Pc was computed to examine the orbital hybridization and electron transport of the catalyst. Additionally, the PDOS of Tm₂-Pc is displayed in Fig. 3. A pronounced peak is observed at the Fermi level (E_f) for Co₂-Pc, signifying metallic characteristics. At this location, the Co 3d peak is stronger than the N 2p peak, indicating that the Co is the main contributor to the conductivity of Co₂-Pc. Mn₂-Pc, Fe₂-Pc, and Ru₂-Pc display tight spacing at the E_f, with the 3d peaks of the transition metals (Tm = Mn, Fe, and Ru) being more pronounced than the 2p peaks of nitrogen, indicating that the conductivity of Tm₂-Pc is predominantly attributed to the transition metals. The Tm central atoms improve the conductivity of Tm₂-Pc, which facilitates charge transfer for the ORR catalytic reaction, as evidenced by the density of states of Mn₂-Pc, Co₂-Pc, Fe₂-Pc, and Ru₂-Pc. The investigation of the structural and binding stability of Tm₂-Pc established a basis for subsequent research on ORR catalysis.

3.2 Analysis of ORR catalytic performance

Following the examination of the structural stability of Tm₂-Pc, the catalytic performance for the ORR was further evaluated. In the four-step ORR reaction, the catalytic efficacy of Tm₂-Pc is contingent upon variations in the free energy of the intermediates (*OOH, *O, *OH).⁵⁵ The free energy of the pathways for Tm₂ (Tm = Mn, Co, Fe, and Ru)-Pc is shown in Fig. 4. Under 0 eV conditions, the free energy diminishes incrementally as the ORR reaction advances. Overpotential is a critical parameter for assessing the catalytic efficacy of the ORR and can be derived for Tm₂-Pc utilizing formula (6). The *O₂ to *OOH transition for Mn₂-Pc has a free energy change of 0.95 eV. Thus, 0.28 eV is the overpotential for the ORR reaction on Mn₂-Pc. In the cases of Fe₂-Pc, Co₂-Pc, and Ru₂-Pc, the rate-limiting step transpires during the conversion of *O₂ to *OOH, exhibiting overpotentials of 0.78 eV, 0.88 eV, and 0.27 eV, respectively. The precious metal Pt demonstrates exceptional ORR catalytic capability, characterized by an overpotential of 0.79 eV.³⁶ The overpotentials of Fe₂-Pc and Co₂-Pc approximate that of Pt; however, the overpotentials of Mn₂-Pc and Ru₂-Pc are markedly lower than that of Pt, rendering them excellent catalysts for the ORR.

Fig. 4 Free energy diagram for the ORR and HER pathways: (a)–(d) Mn₂-Pc, Fe₂-Pc, Co₂-Pc, and Ru₂-Pc.

At an applied potential of 1.23 V, the ORR continues to proceed, without a progressive decline in free energy. The free energy landscape shows simultaneous energy-increasing and energy-decreasing steps. This indicates that energy is necessary to surpass the positive free energy. The free energy for Co₂-Pc gradually rises from *O₂ to *O, whereas the other free energies initially rise and then progressively fall. The d-band center, HOMO–LUMO gap, and density of states are compared. The upward shift of the d-band centre, which increases the Co 3d states at the Fermi level, together with the enlarged HOMO–LUMO gap that limits electron transfer, results in a net decrease in ORR activity for Co₂-Pc.

The hydrogen adsorption energy of Co₂-Pc in the hydrogen evolution reaction (HER) is markedly superior to that of the other Tm₂-Pc, influenced by the d-band center and the HOMO–LUMO gap (Fig. 4). Notably, the ORR and the HER are competing processes in aqueous electrochemical environments.⁵⁶ The performance trends for both reactions, as shown in Fig. 4, can be rationalized by a unified electronic structure perspective. Catalysts like Fe₂-Pc and Co₂-Pc, with their d-band centers positioned closer to the Fermi level, exhibit excessively strong adsorption energies for various intermediates. This is evidenced by their high hydrogen adsorption free energies (ΔG_H) of 0.40 eV and 0.54 eV, respectively. This strong H adsorption not only leads to moderate HER activity but, more importantly, reduces the availability of active sites for O₂ adsorption and the subsequent ORR steps, thereby compromising the overall ORR efficiency. In contrast, Mn₂-Pc and Ru₂-Pc, identified as our top ORR performers, possess lower ΔG_H values (0.25 eV and 0.35 eV, respectively). This indicates weaker H adsorption, which minimizes site-blocking and favors the ORR pathway. This interplay confirms that an optimal electronic structure—one that affords moderate adsorption of ORR intermediates while suppressing competitive H* binding—is key to achieving high ORR selectivity and activity. In conclusion, Mn₂-Pc and Ru₂-Pc exhibit optimal electronic characteristics and appropriate adsorption energies, exhibiting superior electrocatalytic efficacy.

In order to conduct a more thorough examination of the catalytic mechanism of Tm₂-Pc, a volcano plot analysis was implemented. Volcano plots are commonly employed as a method for screening high-performance ORR catalysts. The Tm₂-Pc situated near the summit of the volcano demonstrates superior catalytic performance, but the catalytic performance at the base of the volcano is somewhat worse. Fig. 5(a) illustrates the volcano plot of ΔG_OH* in relation to the overpotential η_ORR. As can be seen, Mn₂-Pc and Ru₂-Pc are both found near the summit of the volcano plot, suggesting low overpotential and superior electrocatalytic activity. In contrast, Co₂-Pc and Fe₂-Pc are found at the middle of the slope, where the ORR catalysis is weaker and the overpotential rises. At the base of the mountain, Pd₂-Pc and Pt₂-Pc exhibit extremely low catalytic activity.

Fig. 5 Volcano curve and 2D volcano diagram: (a) the relationship of −η_ORRvs ΔG_OH*, and (b) the function of the limiting overpotentials and the descriptors of ΔG_O* and ΔG_OH*.

The limiting overpotential analysis is a critical parameter due to the fact that the overall catalytic rate of the reaction is influenced by the stability and catalytic activity of the ORR intermediate (*O). The 2D colored contour volcano plots of ΔG_O* and ΔGOH* were produced based on the correlation between ΔG_OOH* and ΔG_OH* (Fig. S3). Mn₂-Pc and Ru₂-Pc exhibit the lowest limiting overpotentials, signifying a diminished adsorption strength of O*, therefore promoting the spontaneous transformation of the intermediate from O* to OH*, in accordance with Fig. 5(b). The integration of the pathways and contour volcano plots elucidates that the rate-determining step for Mn₂-Pc and Ru₂-Pc is the initial step. For Tm-Pc ORR catalysts, Mn exhibits catalytic efficiency comparable to that of the precious metal Ru, reducing reliance on precious metals and minimizing economic expenditures.

3.3 Analysis of electronic characteristics

The volcano plot examines the catalytic efficiency of Tm₂-Pc by correlating adsorption energy with overpotential. The alteration in electrical structure works as the intrinsic impetus that triggers the response. Consequently, a further thorough study of the electrical characteristics of Tm₂-Pc was undertaken. The projected density of the free O₂ molecule and the Mn atom in Mn₂-Pc was examined (Fig. S4). At the Fermi level, there exists substantial hybridization between the d orbitals of Mn atoms and the p orbitals of O₂. Additionally, the projected density of states for Mn₂-Pc and the schematic representation of the occupancy of the Mn 3d orbitals were investigated (Fig. 6(a)). d–π* orbitals that are partially occupied are the result of the coupling of Mn's d orbitals with O₂'s π* orbitals in the β spin. α spin and β spin have similar patterns. The results demonstrate that O₂ can adsorb onto the Mn₂-Pc substrate, hence promoting the efficient advancement of the ORR reaction. In comparison to Ru₂-Pc, Fe₂-Pc, and Co₂-Pc, the orbital occupancy of Mn₂-Pc and Ru₂-Pc is markedly greater than that of Fe₂-Pc and Co₂-Pc, rendering it more advantageous for the adsorption of adsorbates.

Fig. 6 The influence of the electronic characteristics on ORR catalysis: (a) PDOS and schematic illustrations of the 3d orbitals of Mn₂-Pc and (b) the COHP of Mn and O, and (c)–(e) the differential charge density of OOH*, O* and OH* adsorbed on Mn₂-Pc.

The adsorption ability of Tm₂-Pc on O₂ was investigated by analyzing the COHP between Mn and O atoms. COHP disaggregates the electronic state energy into atomic interactions, measuring bond strength. A negative integrated COHP value (ICOHP) signifies bonding, whereas a positive value denotes antibonding. The magnitude of the number indicates the intensity of the bonding or antibonding interaction. The ICOHP of O₂ adsorbed on Mn₂-Pc is −0.327 eV, signifying that the O₂ molecule is activated and adsorbed by Mn₂-Pc (Fig. 6(b)). The results are consistent with those reported by other researchers,^39,57 further indicating a weakening of the bonding interactions within the O=O bond upon O₂ adsorption. This observation suggests the formation of chemical bonds between O₂ and the Tm₂-Pc substrates. A diminished value enables the release of O atoms from the Mn₂-Pc substrate, improving the sequential stages of the four-step ORR reaction. The intermediates of the four-step ORR (*OOH, *O, *OH) demand charge transfer, and the differential charge quantifies the magnitude of this charge transfer. During the formation of the *OOH, *O, and *OH intermediates, the Mn transfers 0.96 eV, 0.61 eV, and 0.42 eV to the adsorbates, respectively (Fig. 6(c)). Significant charge transfer occurs during the ORR process, indicating an “acceptance–donation” mechanism.^58,59 This mechanism is characterized by charge accumulation on the Tm₂-Pc substrate and concurrent charge depletion on the O atoms. The transfer of charge supplies the energy required for the ORR reaction to take place. The research elucidates the electronic properties of Tm₂-Pc, offering theoretical justification for the synthesis of Tm₂-Pc with superior ORR catalytic performance.

4. Conclusion

Significant challenges remain in the pursuit of clean energy to mitigate the energy crisis and environmental damage. Due to the variety of catalytic sites, Tm₂-Pc offers considerable advantages in ORR electrocatalysis. Nonetheless, the fundamental principles and catalytic processes of Tm₂-Pc remain to be thoroughly investigated. This work utilizes a first-principles approach based on systematic screening to methodically investigate the ORR catalytic processes of Tm₂-Pc. The results indicate that among several Tm₂-Pcs, Mn₂-Pc and Ru₂-Pc demonstrate both structural and binding stability, as well as outstanding ORR catalytic performance, with overpotentials of merely 0.27 eV and 0.28 eV, respectively. This work offers a theoretical foundation and assistance for the development of highly efficient catalysts.

Conflicts of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

All data have been included in the supplementary information (SI). See DOI: https://doi.org/10.1039/d5cp03639a.

Acknowledgements

This research was supported by Yingkou Institute of Technology (YJRC202401), the National Natural Science Foundation of China (52462009, 52402077), and the Funds Project of Liaoning Provincial Engineering Research Center for High-Value Utilization of Magnesite (LMNKY202306).

References

1. J. H. Li, Z. X. Li and Q. H. Sun, Adv. Energy Mater., 2024, 14, 2402441.
2. M. H. Ning, S. N. Wang and J. Wan, Angew. Chem., Int. Ed., 2024, 63, e202415794.
3. M. V. Jyothirmai, R. Dantuluri and P. Sinha, ACS Appl. Mater. Interfaces, 2024, 16, 12437–12445.
4. L. Li, M. Han and P. G. Zhang, ChemSusChem, 2025, 18, e202401186.
5. M. Tamtaji, M. G. Kim and Z. M. Li, Nano Energy, 2024, 126, 109634.
6. A. Kumar, G. Zhang and W. Liu, J. Electroanal. Chem., 2022, 922, 116799.
7. C. X. Huang, G. Li and L. M. Yang, ACS Appl. Mater. Interfaces, 2021, 13, 608–621.
8. Z. J. Liao, Y. L. Wang and Q. L. Wang, Appl. Catal., B, 2019, 243, 204–211.
9. Y. Ma, J. T. Li and X. B. Liao, Adv. Funct. Mater., 2020, 30, 2005000.
10. N. Komba, G. X. Zhang and Q. Wei, Int. J. Hydrogen Energy, 2019, 44, 18103–18114.
11. S. Shi, J. Yang and L. Chen, J. Electrochem. Soc., 2024, 171, 046504.
12. X. Meng, P. Yu and M. Zhang, Molecules, 2024, 29, 4272.
13. F. Rehman, S. Kwon and C. B. Musgrave, Nano Energy, 2022, 103, 107866.
14. G. Hai, X. Xue, Z. Wu and C. Zhang, J. Energy Chem., 2024, 91, 194–202.
15. Q. Zhou, F. Gong and Y. Xie, Green. Energy. Environ., 2023, 8, 1753–1763.
16. Q. Q. An, X. P. Qin and Q. H. Liu, JACS, 2025, 147(13), 11465–11476.
17. Y. Che, J. Shang and X. Chen, At. Force Microsc., 2025, 2507544.
18. S. W. Bo, X. X. Zhang and X. Chen, Nat. Commun., 2025, 16, 2483.
19. S. Yang, Y. Yu and X. Gao, Chem. Soc. Rev., 2021, 50, 12985–13011.
20. Z. Wang, A. Ma and Z. Liu, Appl. Surf. Sci., 2024, 669, 160532.
21. S. A. Mirshokraee, M. Muhyuddin and J. Orsilli, Nanoscale, 2024, 16, 6531–6547.
22. G. Kresse and J. Furthmüller, Comput. Mater. Sci., 1996, 6, 15–50.
23. G. Kresse and J. Hafner, Phys. Rev. B: Condens. Matter Mater. Phys., 1993, 47, 558–561.
24. P. E. Blöchl, Phys. Rev. B: Condens. Matter Mater. Phys., 1994, 50, 17953–17979.
25. G. Kresse and J. Furthmüller, Phys. Rev. B: Condens. Matter Mater. Phys., 1996, 54, 11169–11186.
26. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865–3868.
27. G. Kresse and D. Joubert, Phys. Rev. B: Condens. Matter Mater. Phys., 1999, 59, 1758–1775.
28. S. Grimme, J. Comput. Chem., 2006, 27, 1787–1799.
29. H. J. Monkhorst and J. D. Pack, Phys. Rev. B, 1976, 13, 5188–5192.
30. X. Li, Z. Wang and Z. Su, ChemPhysMater, 2022, 1, 237–245.
31. S. Nosé, J. Chem. Phys., 1984, 81, 511–519.
32. R. Dronskowski and P. E. Bloechl, J. Phys. Chem., 1993, 97, 8617–8624.
33. S. Maintz, V. L. Deringer and A. L. Tchougréeff, J. Comput. Chem., 2016, 37, 1030–1035.
34. G. Henkelman, A. Arnaldsson and H. Jónsson, Comput. Mater. Sci., 2006, 36, 354–360.
35. J. K. Nørskov, T. Bligaard and A. Logadottir, J. Electrochem. Soc., 2005, 152, J23.
36. Y. Li, J. Tan and X. Zhao, Mater. Today Commun., 2023, 37, 107157.
37. M. D. Cueto, P. Ocón and J. M. L. Poyato, J. Phys. Chem. C, 2015, 119, 2004–2009.
38. J. K. Nørskov, J. Rossmeisl and A. Logadottir, J. Phys. Chem. B, 2004, 108, 17886–17892.
39. S. Wang, K. Meng and L. Qin, Int. J. Hydrogen Energy, 2024, 88, 850–857.
40. X. Chen, J. Tan and M. Leng, Int. J. Hydrogen Energy, 2022, 47, 17611–17620.
41. Y. Chen, P. Zhang and X. Liu, Int. J. Hydrogen Energy, 2024, 62, 673–683.
42. X. Y. Guo, J. X. Gu and S. R. Lin, JACS, 2020, 142, 5709–5721.
43. K. H. Yeoh, Y. H. Chang and H. W. Yu, ACS Appl. Electron. Mater., 2025, 7, 3521–3528.
44. D. Watson, Food chemical safety: volume 1: contaminants, Woodhead Publishing, 2001.
45. S. J. Mulware, 3 Biotech, 2013, 3, 85–96.
46. V. Balaram, Geosci. Front., 2019, 10, 1285–1303.
47. L. Lin, Y. Sun and K. Xie, Phys. Chem. Chem. Phys., 2023, 25, 15441–15451.
48. N. Ma, Y. Wang and Y. Zhang, Appl. Surf. Sci., 2022, 596, 153574.
49. L. Ou, F. Yang and Y. Liu, J. Phys. Chem. C, 2009, 113, 20657–20665.
50. S. Jiao, X. Fu and H. Huang, Adv. Funct. Mater., 2022, 32, 2107651.
51. Q. Zhu, W. Huang and C. Huang, Int. J. Hydrogen Energy, 2022, 47, 38445–38454.
52. N. Blankevoort, P. Bastante and R. J. Davidson, ACS Omega, 2024, 9, 8471–8477.
53. Y. Gao, X. Xu and Y. Niu, Nano Res., 2023, 16, 12936–12941.
54. Y. Li, Y. Feng and D. Zheng, Chem. Eng. J., 2023, 476, 146753.
55. I. C. Man, H. Y. Su and V. F. Calle, ChemCatChem, 2011, 3, 1159–1165.
56. P. F. Liu, Z. R. Zhang and X. H. Liu, Int. J. Hydrogen Energy, 2025, 101, 212–221.
57. Y. Yao, S. Y. Lv and G. Li, Phys. Chem. Chem. Phys., 2023, 25, 27131–27141.
58. S. Cao, H. Chen and X. Wei, Carbon, 2024, 225, 119094.
59. P. Xue, X. Zhao and D. Zheng, Comput. Mater. Sci., 2024, 234, 112778.

---